Concurrency: shared objects & mutual exclusion1 ©Magee/Kramer 2 nd Edition Chapter 4 Shared Objects & Mutual Exclusion

Concurrency: shared objects & mutual exclusion2 ©Magee/Kramer 2 nd Edition Shared Objects & Mutual Exclusion Concepts : process interference. mutual exclusion. Models :model checking for interference modeling mutual exclusion Practice :thread interference in shared Java objects mutual exclusion in Java (synchronized objects/methods).

Concurrency: shared objects & mutual exclusion3 ©Magee/Kramer 2 nd Edition 4.1 Interference People enter an ornamental garden through either of two turnstiles. Management wish to know how many are in the garden at any time. The concurrent program consists of two concurrent threads and a shared counter object. Ornamental garden problem:

Concurrency: shared objects & mutual exclusion4 ©Magee/Kramer 2 nd Edition ornamental garden Program - class diagram The Turnstile thread simulates the periodic arrival of a visitor to the garden every second by sleeping for a second and then invoking the increment() method of the counter object.

Concurrency: shared objects & mutual exclusion5 ©Magee/Kramer 2 nd Edition ornamental garden program private void go() { counter = new Counter(counterD); west = new Turnstile(westD,counter); east = new Turnstile(eastD,counter); west.start(); east.start(); } The Counter object and Turnstile threads are created by the go() method of the Garden applet: Note that counterD, westD and eastD are objects of NumberCanvas used in chapter 2.

Concurrency: shared objects & mutual exclusion6 ©Magee/Kramer 2 nd Edition Turnstile class class Turnstile extends Thread { NumberCanvas display; Counter people; Turnstile(NumberCanvas n,Counter c) { display = n; people = c; } public void run() { try{ display.setvalue(0); for (int i=1;i<=Garden.MAX;i++){ Thread.sleep(500); //0.5 second between arrivals display.setvalue(i); people.increment(); } } catch (InterruptedException e) {} } The run() method exits and the thread terminates after Garden.MAX visitors have entered.

Concurrency: shared objects & mutual exclusion7 ©Magee/Kramer 2 nd Edition Counter class class Counter { int value=0; NumberCanvas display; Counter(NumberCanvas n) { display=n; display.setvalue(value); } void increment() { int temp = value; //read value Simulate.HWinterrupt(); value=temp+1; //write value display.setvalue(value); } Hardware interrupts can occur at arbitrary times. The counter simulates a hardware interrupt during an increment(), between reading and writing to the shared counter value. Interrupt randomly calls Thread.sleep() to force a thread switch.

Concurrency: shared objects & mutual exclusion8 ©Magee/Kramer 2 nd Edition ornamental garden program - display After the East and West turnstile threads have each incremented its counter 20 times, the garden people counter is not the sum of the counts displayed. Counter increments have been lost. Why?

Concurrency: shared objects & mutual exclusion9 ©Magee/Kramer 2 nd Edition concurrent method activation Java method activations are not atomic - thread objects east and west may be executing the code for the increment method at the same time. east west increment: read value write value + 1 program counter program counter PC shared code

Concurrency: shared objects & mutual exclusion10 ©Magee/Kramer 2 nd Edition ornamental garden Model Process VAR models read and write access to the shared counter value. Increment is modeled inside TURNSTILE since Java method activations are not atomic i.e. thread objects east and west may interleave their read and write actions.

Concurrency: shared objects & mutual exclusion11 ©Magee/Kramer 2 nd Edition ornamental garden model const N = 4 range T = 0..N set VarAlpha = { value.{read[T],write[T]} } VAR = VAR[0], VAR[u:T] = (read[u] ->VAR[u] |write[v:T]->VAR[v]). TURNSTILE = (go -> RUN), RUN = (arrive-> INCREMENT |end -> TURNSTILE), INCREMENT = (value.read[x:T] -> value.write[x+1]->RUN )+VarAlpha. ||GARDEN = (east:TURNSTILE || west:TURNSTILE || { east,west,display}::value:VAR) /{ go /{ east,west}.go, end/{ east,west}.end}. The alphabet of shared process VAR is declared explicitly as a set constant, VarAlpha. The TURNSTILE alphabet is extended with VarAlpha to ensure no unintended free (autonomous) actions in VAR eg. value.write[0]. All actions in the shared VAR must be controlled (shared) by a TURNSTILE.

Concurrency: shared objects & mutual exclusion12 ©Magee/Kramer 2 nd Edition checking for errors - animation Scenario checking - use animation to produce a trace. Is this trace correct?

Concurrency: shared objects & mutual exclusion13 ©Magee/Kramer 2 nd Edition checking for errors - exhaustive analysis TEST = TEST[0], TEST[v:T] = (when (v TEST[v+1] |end->CHECK[v] ), CHECK[v:T] = (display.value.read[u:T] -> (when (u==v) right -> TEST[v] |when (u!=v) wrong -> ERROR ) )+{display.VarAlpha}. Exhaustive checking - compose the model with a TEST process which sums the arrivals and checks against the display value: Like STOP, ERROR is a predefined FSP local process (state), numbered -1 in the equivalent LTS.

Concurrency: shared objects & mutual exclusion14 ©Magee/Kramer 2 nd Edition ornamental garden model - checking for errors ||TESTGARDEN = (GARDEN || TEST). Use LTSA to perform an exhaustive search for ERROR. Trace to property violation in TEST: go east.arrive east.value.read.0 west.arrive west.value.read.0 east.value.write.1 west.value.write.1 end display.value.read.1 wrong LTSA produces the shortest path to reach ERROR.

Concurrency: shared objects & mutual exclusion15 ©Magee/Kramer 2 nd Edition Interference and Mutual Exclusion Destructive update, caused by the arbitrary interleaving of read and write actions, is termed interference. Interference bugs are extremely difficult to locate. The general solution is to give methods mutually exclusive access to shared objects. Mutual exclusion can be modeled as atomic actions.

Concurrency: shared objects & mutual exclusion16 ©Magee/Kramer 2 nd Edition 4.2 Mutual exclusion in Java class SynchronizedCounter extends Counter { SynchronizedCounter(NumberCanvas n) {super(n);} synchronized void increment() { super.increment(); } We correct COUNTER class by deriving a class from it and making the increment method synchronized: Concurrent activations of a method in Java can be made mutually exclusive by prefixing the method with the keyword synchronized, which uses a lock on the object. acquire lock release lock

Concurrency: shared objects & mutual exclusion17 ©Magee/Kramer 2 nd Edition mutual exclusion - the ornamental garden Java associates a lock with every object. The Java compiler inserts code to acquire the lock before executing the body of the synchronized method and code to release the lock before the method returns. Concurrent threads are blocked until the lock is released.

Concurrency: shared objects & mutual exclusion18 ©Magee/Kramer 2 nd Edition Java synchronized statement Access to an object may also be made mutually exclusive by using the synchronized statement: synchronized (object) { statements } A less elegant way to correct the example would be to modify the Turnstile.run() method: synchronized(people) {people.increment();} Why is this “less elegant”? To ensure mutually exclusive access to an object, all object methods should be synchronized.

Concurrency: shared objects & mutual exclusion19 ©Magee/Kramer 2 nd Edition To add locking to our model, define a LOCK, compose it with the shared VAR in the garden, and modify the alphabet set : 4.3 Modeling mutual exclusion LOCK = (acquire->release->LOCK). ||LOCKVAR = (LOCK || VAR). set VarAlpha = {value.{read[T],write[T], acquire, release}} TURNSTILE = (go -> RUN), RUN = (arrive-> INCREMENT |end -> TURNSTILE), INCREMENT = (value.acquire -> value.read[x:T]->value.write[x+1] -> value.release->RUN )+VarAlpha. Modify TURNSTILE to acquire and release the lock:

Concurrency: shared objects & mutual exclusion20 ©Magee/Kramer 2 nd Edition Revised ornamental garden model - checking for errors Use TEST and LTSA to perform an exhaustive check. Is TEST satisfied? go east.arrive east.value.acquire east.value.read.0 east.value.write.1 east.value.release west.arrive west.value.acquire west.value.read.1 west.value.write.2 west.value.release end display.value.read.2 right A sample animation execution trace

Concurrency: shared objects & mutual exclusion21 ©Magee/Kramer 2 nd Edition COUNTER: Abstraction using action hiding To model shared objects directly in terms of their synchronized methods, we can abstract the details by hiding. For SynchronizedCounter we hide read, write, acquire, release actions. const N = 4 range T = 0..N VAR = VAR[0], VAR[u:T] = ( read[u]->VAR[u] | write[v:T]->VAR[v]). LOCK = (acquire->release->LOCK). INCREMENT = (acquire->read[x:T] -> (when (x<N) write[x+1] ->release->increment->INCREMENT ) )+{read[T],write[T]}. ||COUNTER =

Concurrency: shared objects & mutual exclusion22 ©Magee/Kramer 2 nd Edition COUNTER: Abstraction using action hiding Minimized LTS: We can give a more abstract, simpler description of a COUNTER which generates the same LTS: This therefore exhibits “equivalent” behavior i.e. has the same observable behavior. COUNTER = COUNTER[0] COUNTER[v:T] = (when (v COUNTER[v+1]).

Concurrency: shared objects & mutual exclusion23 ©Magee/Kramer 2 nd Edition Summary  Concepts process interference mutual exclusion  Models model checking for interference modeling mutual exclusion  Practice thread interference in shared Java objects mutual exclusion in Java (synchronized objects/methods).